A5024751017- Machine Fault Diagnosis Techniques
- Fault Detection and Control Systems
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Tensor decomposition and applications
- Structural Health Monitoring Techniques
- Blind Source Separation Techniques
- Image and Signal Denoising Methods
- Laser Design and Applications
- Computational Physics and Python Applications
- Imbalanced Data Classification Techniques
- Plasma Diagnostics and Applications
- Probabilistic and Robust Engineering Design
- Hydraulic and Pneumatic Systems
- Advanced Neuroimaging Techniques and Applications
- Advanced Data Compression Techniques
- Magnetic Bearings and Levitation Dynamics
- Industrial Vision Systems and Defect Detection
- Vibration and Dynamic Analysis
- Rough Sets and Fuzzy Logic
- Welding Techniques and Residual Stresses
- Elevator Systems and Control
- Advanced Adaptive Filtering Techniques
- Sparse and Compressive Sensing Techniques
- Advanced Sensor Technologies Research
Tsinghua University
2020-2025
Guizhou University
2025
Jiangxi Agricultural University
2015-2017
Beijing University of Technology
2011
A novel method is proposed for quantitative and localization diagnosis of a bearing outer ring defect in this paper. This aims to distinguish these defects with different angular positions sizes based on the natural multiples ball spacing. Defects that differ size by spacing will have same time interval, which may lead misdiagnosis. Distinguishing such accurately goal some industries. Therefore, synchronized signal vertical horizontal directions systems are quantitatively analyzed presented...
Abstract Stochastic resonance, recognized as a noise-enhanced signal processing technique, has found extensive applications in fault diagnosis, particularly weak detection. This paper proposes high-dimensional space coupled piecewise bistable stochastic resonance (HSCPBSR) model. approach leverages the advantages of systems to address limitations encountered traditional models. The HSCPBSR model systematically investigates relationship between system parameters and output signal-to-noise...
The localization of outer raceway defect plays a significant role in malfunction elimination, failure cause analysis as well the residual life prediction ball bearings. Based on nonlinear dynamic model for bearing and considering finite size, this article employs detailed mathematical derivation theoretical load distribution system with an located at different angular positions. Therefore, essential mechanism approximate linear relationship between proposed indices, namely...
In this work, the dynamic responses of an uncertain flexible shaft-disk-drum rotor under elastic support and connection are investigated, where parameters considered as unknown but bounded interval variables. The energy method is used to derive deterministic analytical model for rotor, which validated by modal tests. A non-intrusive based on Chebyshev polynomial approximation introduced evaluate response system. Comparative studies Monte Carlo simulation, scanning conducted illustrate...
The traditional high-frequency resonance technology has two disadvantages: (1) frequency (RF) is obtained by using Fourier spectrum (FS) and the bandwidth based on experience, effect often poor due to severe noise interference; (2) separation of compound faults not considered. After systematic research, a feasible fault detection methodology named enhanced adaptive summarized introduced, which basis power (PS) analysis, abbreviated as technique (PS-EAHFRT), realize single bearing. Firstly,...
To address the difficulty in extracting fault features from dual-channel signals, this work proposes a multichannel signal fusion processing method based on Flexible Tensor Singular Spectrum Decomposition (FTSSD) with adaptive embedding dimension selection. Firstly, optimal of trajectory tensor is adaptively determined using proposed Trajectory Dimension Ratio (TDR) index. Once obtained, signals are represented as an tensor. Then, FTSSD employed to decompose and extract feature component...
Addressing the challenges of non-unique decomposition outcomes and prolonged durations in fault feature adaptive extraction algorithm based on tensor decomposition, this paper presents a novel called variable sampling singular spectrum (T-SSD) algorithm. The proposed approach centers decomposing multichannel time series with frequency, leveraging value decomposition. Initially, embedding dimension number resampling points were optimized by power spectral density analysis Subsequently,...
In recent years, large language models have made significant advancements in the field of natural processing, yet there are still inadequacies specific domain knowledge and applications. This paper Proposes MaintAGT, a professional model for intelligent operations maintenance, aimed at addressing this issue. The system comprises three key components: signal-to-text model, pure text multimodal model. Firstly, was designed to convert raw signal data into textual descriptions, bridging gap...
To address the difficulty in extracting fault features from dual-channel signals, this work proposes a multichannel signal fusion processing method based on Flexible Tensor Singular Spectrum Decomposition (FTSSD) with adaptive embedding dimension selection. Firstly, optimal of trajectory tensor is adaptively determined using proposed Trajectory Dimension Ratio (TDR) index. Once obtained, signals are represented as an tensor. Then, FTSSD employed to decompose and extract feature component...
In previous studies, the VMDPgram was creatively proposed by combining variational mode decomposition (VMD) with wavelet packet transform (WPT). Although demonstrates excellent performance in bearing fault diagnosis, there are still some issues that need to be further studied. light of this, this work conducts in-depth studies for unresolved issues. First, view obvious second-order cyclostationarity vibration signal rotating machinery such as bearing, especially presence localized faults,...